Early detection and post-treatment characterization of cancer are key factors in successful cancer therapies. Greater resolution for small cancerous regions would allow earlier detection and would also help in the observation of small tumors after treatment. Now, the primary methods for visualizing cancers are (1) transmission CT, (2) MRI, (3) PET, and (4) SPECT. In this proposal we suggest a new approach to cancer imaging based on radio-labeled antibodies and a Rowland circle x-ray camera. The existing instrument is based on an array of spherically bent, that reflect and focus x-rays by Bragg diffraction. The specific aims of this proposal are [1.] to reconfigure our x-ray spectrometer, multiwire detector, and acquisition software for operation as an imaging instrument at 17 keV, [2.] to conduct a side-by-side comparison of image quality for our instrument vs. a pinhole SPECT camera, using both point sources and phantoms, [3.] to investigate the ultimate spatial resolution that can be achieved with a Rowland circle instrument, [4.] to investigate the most efficient combination of sources and crystals in the 20- 35 keV range, including both x-ray and g-ray sources ( 95Tc, 119Sb, and 125I ) and Si, Ge, and Ta crystals, [5.] to investigate the diffraction angles that provide the best tradeoffs between spatial resolution, energy resolution, and spectrometer efficiency, [6.] based on the results of items 1-5, to design a next generation instrument that can serve a particular niche in cancer imaging. During the course of these studies, we will work closely with chemists, physicists, and radiologists. We hope to make a realistic assessment as to whether an optimized Rowland circle x- ray instrument could serve particular needs better than current technology.